Category [Product announcements](https://claude.com/blog/category/announcements) [Agents](https://claude.com/blog/category/agents) [Enterprise AI](https://claude.com/blog/category/enterprise-ai)Product [Claude Platform](https://claude.com/platform/api)Date August 20, 2026Reading time 5 minShare [Copy link](https://claude.com/blog/computer-use-skills-api-files-api#)https://claude.com/blog/computer-use-skills-api-files-api
计算机操作（Computer use）、技能 API（Skills API）和文件 API（Files API）今天在 Claude 平台上全面开放。计算机操作还新增了一个浏览器操作工具，适用于在 Web 应用中工作的智能体。三者结合，你可以构建能够操作软件、运用团队专业知识并返回成品文件的智能体。
在 Claude 平台上构建智能体
计算机操作让你构建能够操作其所见软件的智能体。给定一张截图，智能体会像坐在键盘前的人一样点击、输入和滚动。这使它能够在那些从未为自动化而构建的应用中工作。新的浏览器操作工具将这一能力扩展到 Web 领域。除了截图之外，智能体还会读取页面结构，并针对特定字段或按钮进行操作，而不是屏幕上的某个位置。
技能 API 和文件 API 让你能够将你的专业知识和文档赋予该智能体。技能是一个包含指令、脚本和模板的文件夹，Claude 仅在任务需要时才加载它。借助技能 API，你可以上传并管理自己技能的版本，然后将它们附加到任何请求中。它们在 Claude 的代码执行沙箱中运行，因此你无需自行托管任何内容。文件 API 是智能体读取和写入文档的存储空间：上传一次 PDF 或电子表格，在后续请求中通过其 ID 引用它，而无需重新发送，并可下载智能体创建的文件。
假设你正在构建一个理赔智能体。它从文件 API 读取受理文档，遵循一个编码了团队申报流程的技能，通过浏览器操作工具在保险公司的 Web 门户中完成提交，并将确认信息保存为文件。已全面开放的代码执行和 Web 搜索也适用于同一流程。
全面开放带来的新变化
- 计算机操作：更新后的计算机操作工具让 Claude 每次轮次可执行多个操作，而不是每次模型调用只执行一个操作，因此任务可以在更少的调用次数和更短的时间内完成。根据我们的 BAA 协议，计算机操作现在也适用于受 HIPAA 监管的工作负载。
- 浏览器操作工具：今天在计算机操作中新增。它使用相同的多操作轮次，并增加了页面结构，因此智能体定位 Web 元素比仅依赖像素更可靠。
- Skills API：一种更简单的 API，用于上传和管理你自己技能的版本。
- Files API：自动文件过期、5 倍更高的速率限制，以及每个组织 1 TB 的存储空间。
“我们的智能体运行在那些没有 API 的医疗和保险系统内部。在新的计算机使用工具上，我们最长的理赔流程从 32 分钟缩短到 13 分钟，在我们测试的每一个流程中，单任务成本下降了约 30%，完成率达到 100%，而且我们的提示词完全没改。”
Davide Locatelli，研究工程师
“Skills API 为我们提供了一种直接的方式，将专门的文档创建能力构建到 Box Agent 中。对一家银行来说，一项技能可以捕捉公司的信贷方法论和已批准的备忘录格式；Box Agent 将其应用于 Box 中已有的财务报表和交易文档，并生成一份有来源依据的信贷备忘录供分析师审阅。银行无需从零构建每个复杂流程，就能获得相应的智能体。”
Matthew Midson，银行业务董事总经理
快速开始
计算机使用工具、浏览器使用工具、Skills API 和 Files API 现已在 Claude 平台上提供。Skills API 和 Files API 也可通过 Microsoft Foundry 获取，更新后的计算机使用和浏览器使用工具即将登陆 Google Cloud 的 Vertex AI。在迁移期间，现有的 beta 集成仍可继续使用。请参阅计算机使用、浏览器使用工具、Skills API 和 Files API 的文档以开始使用。
Category [Product announcements](https://claude.com/blog/category/announcements) [Agents](https://claude.com/blog/category/agents) [Enterprise AI](https://claude.com/blog/category/enterprise-ai)Product [Claude Platform](https://claude.com/platform/api)Date August 20, 2026Reading time 5 minShare [Copy link](https://claude.com/blog/computer-use-skills-api-files-api#)https://claude.com/blog/computer-use-skills-api-files-api
Computer use, the Skills API, and the Files API are generally available on the Claude Platform today. Computer use also adds a new browser use tool for agents that work in web applications. Together they let you build agents that operate software, apply your team's expertise, and return finished files.
Building agents on the Claude Platform
Computer use lets you build agents that operate software they can see. Given a screenshot, the agent clicks, types, and scrolls the way someone at the keyboard would. That lets it work in applications that were never built for automation. The new browser use tool extends this to the web. Alongside the screenshot, the agent reads the structure of the page and acts on a specific field or button rather than a position on screen.
TheSkills API and the Files API let you give that agent your expertise and your documents. A skill is a folder of instructions, scripts, and templates that Claude loads only when a task calls for it. With the Skills APIyou upload and version your own skills, then attach them to any request. They run in Claude's code execution sandbox, so there is nothing for you to host. TheFiles API is storage for the documents an agent reads and writes: upload a PDF or spreadsheet once, reference it by ID in later requests instead of re-sending it, and download the files the agent creates.
Say you're building a claims agent. It reads the intake document from the Files API, follows a skill that encodes the team's filing procedure, completes the submission in an insurer's web portal with the browser use tool, and saves the confirmation back as a file. Code execution and web search, already generally available, fit into the same loop.
What's new with general availability
- Computer use: the updated computer use tool lets Claude take several actions per turn instead of one per model call, so tasks finish in fewer calls and less time. Computer use is also now eligible for HIPAA-regulated workloads under our BAA.
- Browser use tool: new in computer use today. It uses the same multi-action turns and adds page structure, so agents target web elements more reliably than with pixels alone.
- Skills API: a simpler API for uploading and versioning your own skills.
- Files API: automatic file expiration, 5x higher rate limits, and 1 TB of storage per organization.
"Our agents work inside healthcare and insurance systems that have no API. On the new computer use tool, our longest claims workflow went from 32 minutes to 13, cost per task fell about 30% across every workflow we tested, and completion hit 100%, with no changes to our prompts."
Davide Locatelli, Research Engineer
"The Skills API gave us a straightforward way to build specialized document creation into Box Agent. For a bank, a skill captures the firm's credit methodology and approved memo format; Box Agent applies it to the financial statements and deal documents already in Box and produces a source-grounded credit memo for analyst review. Banks get agents for complex workflows without building each one from scratch."
Matthew Midson, Managing Director of Banking
Getting started
The computer use tool, the browser use tool, the Skills API, and the Files API are now available on the Claude Platform. The Skills API and the Files API are also available through Microsoft Foundry, and the updated computer use and browser use tools are coming soon to Google Cloud's Vertex AI. Existing beta integrations keep working while you migrate. See the documentation for computer use, the browser use tool, the Skills API, and the Files API to get started.